20 results about "Random search algorithm" patented technology

The invention relates to the technical field of AGV intelligent transportation, and discloses a lyocell fiber raw material intelligent transportation method and system based on AGV trajectory planning, and the method comprises the steps: obtaining a collision-free path of an AGV from an initial position to a target position through employing a self-adaptive target guidance RRT random search algorithm; deleting redundant nodes of the collision-free path, and fitting the path by adopting an NURBS curve to obtain a fitted path; segmenting the fitting path according to the curvature value of the fitting path and a preset segmentation number; calculating the speed constraint of each segment point in combination with the resultant gravity center and curvature value of the AGV and the carried raw material; and finally, performing acceleration and deceleration planning on each segmented path when the AGV moves along the fitting path by using an S-shaped speed curve to obtain motion control parameters of the AGV, and controlling the AGV to transport raw materials along the fitting path. According to the method, the optimal track planning of the AGV for transporting the Lyocell pulp raw material can be realized, the path accuracy and the transportation stability of the trolley for transporting the large-weight pulp raw material are ensured, and the continuity of Lyocell production is ensured.

The invention discloses an AI-based multi-source information comparison black and grey production detection method, which comprises the following steps: S1, collecting and preprocessing multi-source heterogeneous information data to generate a multi-source information data set; s2, performing semantic coding on the multi-source intelligence data set to obtain an intelligence semantic vector set with unified dimensions; s3, constructing a locality sensitive hash index structure, and generating a preliminary candidate retrieval space; s4, performing screening by using a random search algorithm to generate a candidate comparison sample set; s5, executing a fine comparison operation, and outputting a comprehensive similarity score; s6, black and grey product behavior judgment is carried out based on the comprehensive similarity score; and S7, the high-risk sample is used for updating the Hash bucket index structure. According to the method, semantic vector coding, locality sensitive hashing and random search technologies are fused, so that efficient comparison of multi-source heterogeneous intelligence and intelligent detection of black and grey product behaviors are realized.

The invention discloses a storm surge artificial intelligence forecasting method based on physical equation constraints, and relates to the technical field of artificial intelligence storm surge forecasting. Comprising the following steps: step 1, collecting multi-source data; step 2, assimilating and preprocessing the data; step 3, realizing a machine learning calculation method of physical loss; step 4, constructing a physical and data co-constrained artificial intelligence model; 5, performing hyper-parameter adjustment by using a random search algorithm, and further optimizing the model; and step 6, carrying out interpretability analysis on the forecasting process and result. According to the invention, three control equations for storm surge calculation are introduced into the machine learning model, so that the dependence on the training data volume is greatly reduced compared with a traditional data-driven intelligent forecasting model; and analyzing the change trend of the physical variables in the forecasting process, carrying out visual display and uncertainty analysis, and solving the problem of uninterpretability of an artificial intelligence forecasting model.

The invention discloses a large hydroelectric generator confluence copper ring arrangement optimization method based on a random search algorithm and a greedy strategy. The method comprises the following steps: step 1, establishing a model of a confluence copper ring arrangement scheme; 2, establishing a spatial information model of the confluence copper ring; step 3, constructing constraint conditions of the confluence copper ring; step 4, establishing an optimization objective function; and 5, optimizing a confluence copper ring arrangement scheme by using a random search algorithm in combination with a greedy strategy. The method effectively solves the problems of low manual arrangement efficiency, serious material waste and the like in a traditional copper ring wiring mode, remarkably improves the copper ring confluence efficiency, and is particularly suitable for large hydroelectric generators with huge structures, complex channels and numerous parallel branches.

The invention belongs to the field of soil layer shear wave velocity and damping inversion, and provides a method for joint inversion of a site model based on vertical array seismic records and drilling test data, which comprises the following steps: layering a soil layer, determining initial reference values of shear wave velocity and damping ratio after layering, and determining an observation transfer function. Constructing an inversion objective function and performing iterative search on an optimal solution by using a hill-climbing-random search algorithm, wherein the key point is to construct the inversion objective function and perform iterative search on the optimal solution by using the hill-climbing-random search algorithm; according to the method, the field model is inversed by using the actually measured vertical station seismic record and the drilling test data in a combined manner, the problem of multiplicity of solutions of a traditional inversion method is effectively solved, the shear wave velocity and damping ratio structure obtained through inversion can effectively represent the amplification effect of a local field on seismic oscillation, meanwhile, the drilling test result can be reflected, and the method is suitable for the field simulation. Therefore, the method accords with reality.

The application discloses a storm surge artificial intelligence prediction method based on physical equation constraints and relates to the technical field of artificial intelligence storm surge prediction, and comprises the following steps: step 1, collecting multi-source data; step 2, data assimilation and preprocessing; step 3, realizing a machine learning calculation method of physical loss; step 4, constructing an artificial intelligence model jointly constrained by physics and data; step 5, using a random search algorithm to adjust hyperparameters, and further optimizing the model; and step 6, performing explainability analysis on a prediction process and results. The three control equations used for storm surge calculation are introduced into the machine learning model, the dependence on the training data volume is greatly reduced compared with a traditional data-driven intelligent prediction model, the trend of the change of the physical variables in the prediction process is analyzed, visual display and uncertainty analysis are performed, and the unexplainability problem of the artificial intelligence prediction model is solved.

The invention discloses a method and a system for realizing multispectral switching by cooperatively regulating and controlling pumping and polarization states, and belongs to the technical field of ultrafast fiber lasers and automatic control. The method comprises the following steps: inputting an output signal of an ultrafast fiber laser into a residual neural network for identification and obtaining an ensemble solution of voltage and pumping power corresponding to a spectrum; searching optimal set solutions corresponding to different spectrum states by using a self-adaptive gradient optimization algorithm; a search result is fed back to an electric control polarization controller and a pump, so that automatic switching of spectrum states is realized; if it is monitored that the spectrum state disappears, an adaptive recovery algorithm is executed, and random extraction is carried out in the set solution for recovery; and if recovery fails, searching through a random search algorithm until recovery. The problem that a passive mode-locking fiber laser based on nonlinear polarization rotation cannot rapidly switch the pulse state and is difficult to stabilize is solved, meanwhile, full-automatic control over the fiber laser is achieved, the universality of an algorithm is improved, and the structure is simple and easy to achieve.

The invention relates to the technical field of protection engineering, in particular to a facility system first-aid repair resource allocation method for a complex uncertain attack scene, which comprises the following steps of: S1, acquiring a facility system network structure and a facility number, and evaluating a facility initial function; s2, determining the level of an adverse event; s3, determining facility system first-aid repair resource preset planning constraint conditions, and constructing an improved facility system first-aid repair resource preset planning model; and S4, solving the first-aid repair resource preset planning model by adopting an adaptive random search algorithm (ARS-SVM) fused with a support vector machine to obtain an optimal solution of the first-aid repair resource configuration of the facility system. According to the method, the resource demand satisfaction degree expectation which maximally considers the conditional value-at-risk is taken as a target, the demand is regarded as a random variable, the probability distribution is obtained through historical data fitting, and the auxiliary variable is introduced to quantify the demand fluctuation, so that the urgent repair recovery accuracy can be improved, the cost effectiveness can be balanced, and the system adaptability can be enhanced.

The present invention discloses a method for predicting and controlling mutations in an online group opinion evolution system, belonging to the technical fields of text mining and mathematical modeling. By processing network text data, system variables of group opinion evolution are obtained; and a mutation model of group opinion evolution is established based on catastrophe theory; an elasticity index model of the system is constructed from the mutation model to analyze the possibility of opinion mutation and the danger of system state, thereby predicting mutations. A method for finding the threshold of the system elasticity index is proposed based on the elasticity index model, and opinion mutations are prevented according to the comparison result between the elasticity index and the threshold. Furthermore, a control method for the elasticity index is proposed, and a random search algorithm for reducing the system elasticity index is proposed to reduce the possibility of system mutations. Thus, the present invention can accurately describe and predict the mutation phenomenon of the online group opinion evolution system, and can reduce the possibility of group opinion mutations with less time consumption.

The present invention discloses a ship motion response prediction method based on Gaussian interpolation, which relates to the technical field of ship motion response prediction, and includes the following steps: S1, data acquisition and preprocessing, including data cleaning, coordinate system conversion, data standardization and data set division; S2, constructing a Gaussian interpolation model, including setting an initial σ value and selecting interpolation points; S3, optimizing model parameters by a random search algorithm, including calculating the weight of each point in a training set, model verification, and model parameter optimization; S4, ship motion response prediction, substituting the ship to be predicted and sea state data into the Gaussian interpolation model to predict the ship motion response. The present invention uses a Gaussian interpolation model to accurately simulate the motion response of a ship under complex sea conditions. Compared with traditional methods, the Gaussian interpolation model can more accurately depict the complex relationship between sea state parameters and the ship motion state with its excellent fitting ability for multidimensional and nonlinear data.

The present invention relates to a method, device and storage medium for solving cold chain transportation optimization with processing time, which is applied to the field of cold chain logistics technology. The method comprises: establishing a route optimization model and limiting the constraints of the model, solving the problem with the minimum sum of production and transportation time and the maximum sum of product profits as constraints, obtaining the initial solution of the population through a random search algorithm, and setting irrelevant solutions in the initial solution as non-dominated solutions; optimizing the solution through the optimization coefficient. The solution adopts some ideas of the differential evolution algorithm, combines the characteristics of discrete optimization and multi-objective optimization, and proposes an optimization coefficient with the characteristics of good convergence and simple implementation. When various constraints are met, the transportation and production plans are automatically calculated and generated, thereby achieving a balance between production time and profit, and providing managers with comprehensive decision support, which is no longer limited to transportation efficiency.

Various embodiments relate to a method of testing a plurality of devices of the same type wherein each of the plurality of devices of the same type include a built-in self-test device, including: randomly generating, by a processor, stimulus parameters; applying, by the built-in self-test devices, the generated stimulus parameters N times to the plurality of devices of the same type; measuring, by the plurality of devices of the same type, a response of the plurality of devices of the same type to the generated stimulus parameters to produce M×N response outputs, where M is a number of the plurality of devices of the same type; calculating, by the processor, a defect likelihood for a test set of the plurality of identical devices based upon a mean of a reference set of the plurality of identical devices response outputs, a mean of the test set response outputs, a standard deviation of reference set response outputs, and a standard deviation of the test set response outputs; determining, by the processor, that the defect likelihood for the test set is greater than a first threshold value; applying, by the processor, an initial step of a directed random search algorithm to update stimulus parameters in response to determining that the defect likelihood is greater than the first threshold; applying, by the built-in self-test devices, the updated stimulus parameters N times to the plurality of devices of the same type; measuring, by the plurality of devices of the same type, a response of the plurality of devices of the same type to the updated stimulus parameters to produce M×N updated response outputs; calculating, by the processor, a defect likelihood for the test set based upon a mean of the reference set updated response outputs, a mean of the test set updated response outputs, a standard deviation of reference set updated response outputs, and a standard deviation of the test set updated response outputs; and determining, by the processor, that the defect likelihood for the test set is greater than a second threshold, wherein the second threshold is greater than the first threshold.

The invention discloses a multi-lake model automatic calibration method based on a random search algorithm. The method comprises the following steps: S1, data preparation and operation environment initialization; s2, parameter space construction and search control setting; s3, candidate parameter combination generation; s4, parameter writing and multi-model operation; s5, output analysis and time sequence alignment are carried out; s6, performance evaluation and effectiveness judgment; s7, performing global random search and optimal candidate screening; s8, local refined searching and optimal parameter updating are carried out; s9, summarizing results and determining final parameters; according to the method, unified initialization, parameter writing, synchronous operation and analysis and evaluation record output are performed on four lake models including FLake, GLM, GOTM and Simstrap under the same framework, manual configuration and repeated trial and error are remarkably reduced, and calibration efficiency and reproducibility are improved.

The present invention belongs to the technical field of UAV swarms, and proposes a method for distributed cooperative target search of UAV swarms. The implementation steps are as follows: (1) Construct a distributed cooperative target search scenario for the UAV swarm and initialize parameters; (2) Establish a search cost function for each UAV; (3) Let the decision time k = 1, and let the pheromone content of the target presence information; (4) Each UAV solves the optimal control input matrix based on a distributed stochastic search algorithm integrating genetic operations; (5) Obtain the distributed cooperative target search result of the UAV swarm. In the present invention, at each decision time, the optimal control input matrix is solved based on a distributed stochastic search algorithm integrating genetic operations. Each UAV not only considers the solution that maximizes its own interests, but also considers other solutions that may bring greater swarm interests, effectively improving the search efficiency of the UAV swarm.

The invention particularly relates to a semiconductor wafer manufacturing system scheduling method based on an improved genetic algorithm, and the method comprises the steps: building a time Petri net model according to a wafer manufacturing system and a processing flow, and determining an initial identifier, a target identifier and a transition emission rule of the time Petri net model; initializing parameters of a genetic algorithm, and generating an initial population by adopting a random search algorithm; the fitness function of the individuals in the population is calculated, the fitness function is updated through a fitness sharing mechanism, and the fitness sharing mechanism comprises the steps that the similarity between the individuals is calculated, and the fitness value is adjusted to maintain population diversity; and performing selection, crossover and mutation operation on the population according to the updated fitness function, and performing iteration to obtain a scheduling result of the time Petri network model. On the basis of the improved genetic algorithm, the exploration space of the Petri network model reachable graph is expanded through a shared fitness mechanism, so that a local optimal solution is jumped out to obtain a better scheduling result.

According to various embodiments, the disclosure discloses systems, methods and media for formulating training datasets for learning-based components in an autonomous driving vehicle (ADV). In an embodiment, an exemplary method includes allocating training datasets for training a learning-based model in the ADV, each training dataset being allocated to one of multiple predefined driving scenarios; determining a weight of each training dataset out of the training datasets; and optimizing the weight of each training dataset in one or more iterations according to a predetermined algorithm until a performance of the learning-based model reaches a predetermined threshold. The predetermined algorithm is one of a random search algorithm, a grid search algorithm, or a Bayesian algorithm.

The application discloses a pilot optimization method based on a combination of a genetic algorithm and a random search algorithm for an OTFS system, and the method comprises the following steps: initializing a population; calculating the fitness value of each individual in the population; entering a breeding cycle iteration; performing a cross-recombination operation on offspring; performing a mutation operation on the offspring; performing a reinsertion operation on the offspring; selecting excellent individuals and performing a random search algorithm on the excellent individuals; aiming at a pilot optimization problem in channel estimation of an orthogonal time frequency space (OTFS) system based on compressed sensing, taking minimization of a cross-correlation value of a recovery matrix as a target, and taking the minimum cross-correlation value as a criterion for selecting the best pilot. The method has the advantages of the genetic algorithm, and a part of the population is randomly searched to avoid falling into a local optimum problem. Compared with a conventional genetic algorithm, the method has better effect, and a pilot obtained by using the method can obtain a lower channel estimation mean square error when applied to channel estimation.

The embodiment of the invention discloses a parameter optimization-based wind power plant single-machine aggregation method and device, equipment and a medium, and the method comprises the steps: carrying out the single-machine equivalence of a target wind turbine generator cluster in a wind power plant, and obtaining an equivalent unit of the target wind turbine generator cluster; performing aggregation according to all the equivalent units to obtain a doubly-fed fan simulation model of the wind power plant; searching the control parameters of the doubly-fed fan simulation model based on a grid random search algorithm to obtain the control parameters of the doubly-fed fan simulation model; and configuring a doubly-fed fan simulation model according to the control parameters so as to realize aggregation of equivalent units. The method comprises the following steps: performing single-machine equivalence on each wind turbine generator cluster to obtain a corresponding equivalent unit, aggregating the equivalent units to obtain a doubly-fed fan simulation model, and obtaining an optimal solution of control parameters of the doubly-fed fan simulation model through a grid random search algorithm to achieve the best fitting effect of the doubly-fed fan simulation model. The simulation model has the advantage of being good in simulation effect.

The application provides a logistics route fusion method and device, electronic equipment and storage medium. The logistics route fusion method comprises the following steps: acquiring at least one first to-be-fused line corresponding to a target original line in a first logistics network and at least one second to-be-fused line in a second logistics network; pairing and fusing the at least one first to-be-fused line and the at least one second to-be-fused line by using a random search algorithm to obtain at least one fused line, wherein the fused line is a cost-time optimal to-be-fused line in a plurality of to-be-fused lines corresponding to each second to-be-fused line; determining the at least one fused line as a first fused line combination of the target original line; calculating the first fused line combination by taking each original line in the first logistics network as the target original line to obtain a plurality of first fused line combinations corresponding to the plurality of original lines; and outputting logistics line fusion information based on the plurality of first fused line combinations.

This invention discloses a sparse antenna array for a vehicle-mounted 4D millimeter-wave radar and its design method. The radar system used in this invention includes 12 transmitting antennas and 16 receiving antennas, which cannot overlap. The 12 transmitting antennas are divided into two groups, TX1 and TX2, where TX1 consists of 9 antennas and TX2 consists of 3 antennas. The 16 receiving antennas are also divided into two groups, RX1D and RX2D. TX1 and RX1D are combined to form a one-dimensional virtual array for two-dimensional target detection, and TX1 and RX2D are combined to form a two-dimensional virtual array for three-dimensional target detection, providing azimuth and elevation detection capabilities. This invention utilizes a random search algorithm to successively optimize the main lobe width and side lobe level of the radiation patterns of the two-dimensional and one-dimensional virtual arrays, improving the angular resolution in both the elevation and azimuth directions without increasing the radar's elevation dimension.